tech-pkg archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: reducing pkg PRs

> "Enable PKG_DEVELOPER=yes. Do not disable checks. Test before commit"
> I don't see any point in explaining the rather long list. The other
> defaults are generally the way they are because they tend to hit too
> many false positives.

What kind of "false positives" you are talking about?  As I said my
Linux bulk builds set PKG_DEVELOPER=yes for all packages, and I don't
remember any false positives at all. In the last bulk build 7728
packages were built successfully, more than on, say, NetBSD/spark.  Most
of the rest is fixable is doesn't look like "too much".

Can you point me to "too many false positives"?

The following is from private discussion with Joerg. He gave me rights
to copy this here.

>>> Bulk builds are a separate issue.
>> Bulk builds is the only way to improve pkgsrc packages globally
>> including support for other platforms.  "Use PKG_DEVELOPER=yes locally"
>> approach obviously doesn't work and will not.

>I don't disagree.

Don't or do?

>>> I have complained more than once about the issue of not using
>>> PKG_DEVELOPER for those
>>I probably overlooked them. Url?

>No idea if those have been on public lists.

Could you please list these issues once again, now in public list?
After many year of using pkgsrc I still don't understand
why you don't set PKG_DEVELOPER=yes in official bulk builds.

Best regards, Aleksey Cheusov.

Home | Main Index | Thread Index | Old Index